Hi Jason,

> Nessus will not tell you anything about a router's security. It'll
> tell you a lot about the presence of already known vulnerabilities,
> but nothing about how secure (or better, insecure) such a router is.

I agree, that Nessus can only find known vuls, but to me 994 plugins are an 
awful lot and I feel, that if a router scores good in that test it�s an important 
piece in the puzzle.
Sure, if a user doesn�t change the default admin password, how strong the 
router holds up against other attacks.

> > Is this a good idea that will yield relevant rsults? Did I miss
> > something? Any suggestions for improvments?
> 
> You missed the pen-test side of it. Vulnerability enumeration is but a
> small part of the pie.
Ok then, what�s the pen-side for you? I thought, that what Nessus is doing is 
pen-testing. It tries all the buffer overflows and cross side scripting + the 
nmap-scanning. 
Sure, there is no individual sitting there trying to get into the machine, but pls 
bear in mind: 
We�re talking about SOHO-routers, so the router must be configured pretty 
decent, when it comes to security, but I belive that only very few of my 
readers - if any - will face the situation, that there�s someone, that under all 
circumstances will break into their router. I think, that if a router fends off the 
kiddies and their scripts + all unneccessary services closed + strong 
passwords it should be fine for the target group.

But by any means: Tell me, what else I could/should do - the better the article 
gets the happier I am.

> It's - to mention something we are both familiar with :) - like
> running Nessus against a Zope server. Or, even worse, one of the other
> commercial products. None of them knows about Hotfixes - simply
> because Zope has such a small 'Marketshare' (it still rocks, though).

But has already it�s own Nessus plugin :-)

> How abot weak passwords, predictable passwords, backdoors? Nessus does
> not address them, and was never meant, to.

> Netopia, FlowPoint, Cayman, they all have and had their share of bugs
> and vulnerabilities, most of them undetectable through simple
> vulnerability enumeration but easily found through diligent research
> and a good portion of curiosity applied to common knowledge.
> 
> Let me not be misunderstood here - Vulns found by Nessus are a serious
> concern, simply because they're the most likely ones to be found and
> exploited. There's gazillions of twelve-year-olds out there who have
> nothing better to do than run Nessus against random netblocks after
> school, but that's only a small fraction of what pen-tests are all
> about.

> jonas


Reply via email to